The following parameters can be used in RFC client programs and RFC server programs alike.
Name NAME Each destination and
server needs to be given a name. The application can then access the
destination/server via this name N by calling RfcDestinationMana
PartnerCharSize PCS "Partner character size".
In 99.9% of the cases you don't need to bother with that. During the initial handshake, NCo obtains the correct value from the backend and uses it from then on.
One rare use case is as follows: you know that the backend is Unicode and want to use a non-ISO-Latin-1 username or password for the initial logon. As the initial handshake is done with ISO-Latin-1, the characters in username/passwd would break, resulting in a refused logon. In that case set PCS=2 and NCo will use Unicode for the initial handshake.
1: Non-Unicode 2: Unicode
Codepage CODEPAGE Similar to PCS above.
You only need it if you want to connect to a non-Unicode backend using a non-ISO-Latin-1 username or password.
NCo will then use that codepage for the initial handshake, thus preserving the characters in
username/password.
Another use case is, if you logon to a backend, whose “system
codepage” is different from 1100, and an error happens during the initial handshake before the partners can
exchange their supported codepages.
The backend will return an error message in its system codepage (e.g.
Japanese Shift-JIS) and NCo will try to interpret the error message as 1100, making it illegible.
A few common values are:
1401: ISO-Latin-2 However, please note that these values can be customized in the backend.
Better consult the backend sysadmin first, as otherwise things may go terribly wrong.
OnCharacterConversionError ON_CCE “Character Conversion Error”
What shall NCo do when it encounters a character that does not
This parameter can take three values:
0: Abort with an error message (default behavior).
Note that in this case
exist in the target codepage, a broken character, or a control character (0x00 - 0x19)?
control characters (e.g.
tabulator, carriage return, or linefeed characters) are not considered "illegal" and will therefore not cause an abort.
1: Copy the character in a
"round-trip compatible way". The resulting output character may be "garbage"
in the target codepage, but when converted back to the source codepage, it will be the original character.
2: Replace the character with a substitute symbol (usually a # character).
Note that in this case the control characters are replaced as well. If you need the control characters, then you'll have to use option 0 or 1, depending on whether you want the NW RFC Lib to abort the call in case of broken characters or not.
CharacterFaultIndicatorToken CFIT “Conversion Fault Indicator Token"
The substitute symbol used if ON_CCE=2.
Needs to be given as hexadecimal value of a Unicode codepoint.
The default is 0x0023 ("#
character").
UseSAPCodepages USE_SAP_CODEPAGES By default, NCo uses Microsoft’s codepage converters contained in the .NET Framework.
However, for certain SAP Codepages this may lead to incorrectly translated and broken characters, for example, if the backend is running a “blended codepage” or an East-Asian
codepage.
In this case you can specify a list of codepages for which you want NCo to use the SAP codepage converters.
Comma-separated list of SAP codepages, e.g.
6100,8000,8340, or * for all.
RepositoryDestination REPOSITORY_DESTINA TION
Provides the name of a destination that should be used for retrieving metadata information (function module and
The “Name” of a destination
structure descriptions).
For a destination, you will not need this in most of the cases. Usually a destination uses itself for the necessary DDIC lookups. However, if you are communicating with a large number of different backend systems, all of which have the same release (and consequently identical metadata information), it would be wasteful to cache that identical metadata information several times. Instead you can use only one system and its corresponding DDIC cache and assign this system to the remaining n-1 destinations as a
“repository destination”.
Then we store all DDIC information in memory only once, and the other destinations reuse the information from the first destination.
Instead you can also set the repository
programmatically, for example when using a hard-coded repository (RfcCustomReposito ry).
Trace TRACE Sets a trace level for
this specific destination or server.
Same values as for the default trace level.
SncMode SNC_MODE Determines whether
connections will be secured with SNC.
0: do not use SNC (default) 1: use SNC
SncMyName SNC_MYNAME In most cases this can
be omitted. The installed SNC solution usually knows its own SNC name. Only for solutions supporting “multiple identities”, you may
Varies depending on the installed SNC solution (Secude, Kerberos, NTLM, etc).
Example for Secude:
p/secude:CN=ALEREMOT
need to specify the identity to be used for this particular
destination/server.
E, O=Mustermann-AG, C=DE
SncPartnerName SNC_PARTNERNAME The backend’s SNC
name.
See SncMyName
SncLibraryPath SNC_LIB Full path and name of
the SNC shared library to be used.
SncQOP SNC_QOP Quality of Service to be
used for SNC
communication of this particular
destination/server.
One of the following values:
1: Digital signature 2: Digital signature and encryption
3: Digital signature, encryption, and user authentication
8: Default value defined by back-end system
9: Maximum value that the current security product supports
SAPRouter SAPROUTER If the connection needs
to be made through a firewall via a
SAPRouter, specify the SAPRouter parameters here.
A list of host names and service names / port numbers in the following format:
/H/hostname/S/portnumber
NoCompression NO_COMPRESSION By default the RFC
protocol compresses tables when they reach a size of 8KB or more.
On very rare occasions you may want to turn this off, for example if you are transporting huge integer/binary tables with "random"
data, where compression would have no notable effect except for wasting CPU cycles. Or if you are trouble-shooting a certain problem and want to see the table in the trace file in human-readable format.
0: Compress tables (default)
1: Do not compress tables